Computer Vision News - May 2022

23 AI Spotlight News An Optimized Solution for Face Recognition More News from MIT. They’re good! When AI is tasked with visually identifying objects and faces, it assigns specific components of its network to face recognition - just like the human brain. Which means that it’s a specific area of our brain that enables each of us to recognize hundreds or thousands of people’s faces; and that AI trained to identify faces and other objects discovers a surprisingly brain-like strategy to sort them all out. What stroke MIT’s McGovern Institute for Brain Research scientists is that two completely different systems have independently devised a very similar solution to the same problem. Read More Israeli-Based Tech Startup Brings Your Old Family Photos to Life with Amazing Artificial Intelligence This Forbes article relates of something amazingly beautiful and touching. Leveraging Artificial Intelligence and sophisticated technology, the company has created a unique, animated, live portrait, which animates the photos of long-lost relatives or whoever you’d like to see, as if they are in the room with you. Its tech makes people come alive and look realistic and natural. The feature, Deep Nostalgia , lets users upload a photo of a person or group of people to see individual faces animated by AI, breathing life into old black-and-white photos of grandma and grandpa stored in boxes up in the attic. Read More A New State of the Art for Unsupervised Computer Vision That would sound too ambitious, if it didn’t come from the guys at MIT CSAIL : they created an algorithm to solve one of the hardest tasks in computer vision: assigning a label to every pixel in the world, without human supervision. “STEGO” can jointly discover and segment objects without any human labels at all, down to the pixel. To discover these objects without a human’s guidance, STEGO looks for similar objects that appear throughout a dataset. It then associates these similar objects together to construct a consistent view of the world across all of the images it learns from. Watch the Video